Artificial Living Brain Laboratory
The Artificial Living Brain Laboratory is a research lab pioneering the development of next-generation Neural Processing Units (NPUs) for autonomous robotic systems. Our mission is to create radically energy-efficient NPUs, optimized for mobile platforms.
Our Vision: Artificial Living Brains
To overcome the limitations of traditional silicon-based chips, we are engineering novel non-biological artificial living materials that replicate the functionality of biological neural networks. These materials enable the construction of “living” processors—chips capable of advanced computation, adaptive behavior, and cognitive complexity, while consuming a fraction of the energy of conventional NPUs.
Mobile robots powered by these artificial living brains will be able to operate autonomously in complex, dynamic environments, making real-time decisions without external computation.
We combine artificial life and artificial intelligence to create a new kind of neural processing units (NPUs) composed of artificial living materials that replicate the functionality of biological neural matter. These “artificial living brains” represent a major leap forward, with the potential to significantly outperform existing NPUs in terms of speed, processing capacity, and cognitive complexity.
What is an NPU?
A Neural Processing Unit (NPU)—also known as an AI accelerator—is a specialized microprocessor designed to perform the massive parallel computations required by modern artificial intelligence algorithms. Companies like NVIDIA, a leader in AI chip development, have seen massive growth, with Nvidia becoming the most valuable semiconductor company in history.
As the demand for intelligent systems grows, the next generation of NPUs will need to go beyond conventional architectures. New designs will unlock entirely new markets and applications—particularly in energy-constrained, mobile, or embedded systems.
Our approach
Our Laboratory’s approach is grounded in a deep integration of physics, materials science, and cognitive computation. We develop non-biological neural materials that exhibit life-like information processing. Using fabrication techniques inspired by semiconductor manufacturing, we ensure compatibility with existing microelectronics infrastructure.
We collaborate with top academic and research institutions that conduct fundamental and applied research in artificial life, physics of neural networks, and cognitive sciences. All proprietary materials, fabrication methods, and processing architectures remain the intellectual property of the Animate Condensed Matter Company.
Overcoming the Complexity Barrier
The biological brain is the most complex system known, with countless interdependent variables—making it impossible to model precisely using conventional mathematics.
Rather than attempting to design these systems outright, we take a different path: experimental evolution. We fabricate many simple neural specimens. We test them in real-world and virtual environments. We select and replicate the most successful variants in increasingly complex environments. Over successive generations, complexity and functionality evolve naturally.
This process, akin to the selective breeding of genetically modified organisms, allows us to grow intelligence rather than engineer it from scratch.
Living Brain Ecosystems in Virtual Environments
Not all artificially evolved brains are safe. Many will display undesirable behaviors. For this reason, the development of these systems must occur first in high-fidelity virtual environments, where risks can be isolated and controlled.
In these virtual worlds—similar to The Matrix—artificial creatures evolve, learn, and adapt. Here, humans act as overseers and moderators of complex emergent behaviors. Crucially, the success of such an ecosystem depends on scale and diversity. To accelerate this evolution, we need massive infrastructure: research facilities, compute resources, robotic platforms, and international collaboration.
A Call to Action
We call upon global technology companies, government agencies, the scientific community, and all sectors in need of high-skill labor, to begin building this infrastructure now. This is your opportunity to gain decisive competitive advantage. An intelligent machine workforce can reinforce domestic manufacturing, drive sustained economic growth, unlock new territories and natural resources, and ultimately ensure political and military supremacy.
How the living brains are made
From a physics perspective, the living cell is a complex macromolecule, in which collective chemical transformations drive orderly microscopic movements. These precise movements enable the cell to extract nutrients from its environment, transport essential materials to specific locations, synthesize necessary products, and integrate them into the cell’s structure. At its most fundamental level, the living organism is focused on self-renewal, replacing used components with fresh ones.
Many non-biological systems exhibit similar self-sustaining processes. Some of those can be used to construct artificial cells. A key challenge is that both biological and artificial cells struggle to adapt to dynamic environments. They are more suited to stable conditions, as extreme structural or metabolic changes can lead to the organism’s demise.
Multicellular animals solve this problem through an adaptive organ —the brain. The brain contains neurons that form numerous dynamic connections with each other. These neurons remain in place but constantly modify their connections, forming new patterns and structures. Neurons’ ability to reconfigure their connections enables the brain to perform a wide range of complex actions. Additionally, neurons can prune old connections and grow new ones, allowing the brain to forget outdated behaviors and learn new ones.
AI accelerators: the Building Blocks of AI
To replicate the adaptability of biological brains, researchers have developed the NPUs. These electronic chips are multi-core processors, with each core acting as an electronic analog of a neuron. While the structure and function of each core remain fixed, the connections between the cores can be modified, allowing the system to learn and adapt. These dynamic connections are stored in dedicated memory cells within the NPUs, enabling the system to continuously learn new information.
As the total number of NPUs in AI systems increases, so does their cumulative capability. To continue advancing AI scientists must reduce their energy consumption and explore new ways to process information. This is where innovative new types of NPUs come into play. Our lab is at the forefront of this exploration.